Re: drm/i915: Add DSC/FEC support info to debugfs

[Jani Nikula <jani.nikula@linux.intel.com>]

On Thu, 03 Apr 2025, Imre Deak <imre.deak@intel.com> wrote:
> On Fri, Apr 04, 2025 at 12:34:48AM +0530, shantam.yashashvi@intel.com wrote:
>> From: Shantam Yashashvi <shantam.yashashvi@intel.com>
>> 
>> Signed-off-by: Shantam Yashashvi <shantam.yashashvi@intel.com>
>> ---
>>  drivers/gpu/drm/i915/display/intel_display_debugfs.c | 11 +++++++++++
>>  1 file changed, 11 insertions(+)
>> 
>> diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs.c b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
>> index 057a8ce0c2c97..b8d3a438a34d9 100644
>> --- a/drivers/gpu/drm/i915/display/intel_display_debugfs.c
>> +++ b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
>> @@ -276,6 +276,7 @@ static void intel_connector_info(struct seq_file *m,
>>  	struct intel_encoder *encoder =
>>  		to_intel_encoder(conn_state->best_encoder);
>>  	const struct drm_display_mode *mode;
>> +	struct intel_dp *intel_dp;
>>  
>>  	seq_printf(m, "[CONNECTOR:%d:%s]: status: %s\n",
>>  		   connector->base.id, connector->name,
>> @@ -301,6 +302,16 @@ static void intel_connector_info(struct seq_file *m,
>>  			intel_dp_mst_info(m, intel_connector);
>>  		else
>>  			intel_dp_info(m, intel_connector);
>> +
>> +		/* Add DSC and FEC Support Information for DisplayPort / eDP Connectors */
>> +		intel_dp = intel_attached_dp(intel_connector);
>> +		if (intel_dp){
>> +			seq_printf(m, "\tDSC_Sink_Support: %s\n",
>> +                str_yes_no(drm_dp_sink_supports_dsc(intel_connector->dp.dsc_dpcd)));
>> +			if (!intel_dp_is_edp(intel_dp))
>> +				    seq_printf(m, "\tFEC_Sink_Support: %s\n",
>> +				        str_yes_no(drm_dp_sink_supports_fec(intel_connector->dp.fec_capability)));
>
> Both of the above info is available already in the connector's
> i915_dsc_fec_support debugfs entry. As I understood the aim is having
> each CRTC / connector show its own properties, instead of combinining
> all those into one debugfs entry (as this is done in i915_display_info).
> Based on that I don't think more (duplicated) information should be
> added here.

Agreed. I think the direction should be to reduce stuff from
i915_display_info and put it into per-crtc and per-connector debugfs
instead.

BR,
Jani.



>
> One issue with the i915_dsc_fec_support entry is that it doesn't show
> the sink capabilities if the output isn't enabled on the connector, I
> plan to follow up with a fix for that.
>
>> +		}
>>  		break;
>>  	case DRM_MODE_CONNECTOR_HDMIA:
>>  		if (encoder->type == INTEL_OUTPUT_HDMI ||
>> -- 
>> 2.25.1
>> 

-- 
Jani Nikula, Intel